The purpose of this R21 application is to develop an automated, chip-based metabolomic analysis for point of-practice applications. Intracellular metabolite measurements in current forms are tedious, inaccurate, and expensive. However, their importance has grown significantly as various high-throughput technologies for genomics, transcriptomics, and proteomics have been developed. The measurement of metabolite profiles becomes the next challenge before integrating various "omic" data to explain complex physiology at the systems level. Typically, intracellular metabolite measurements require extensive operations for quenching the metabolism and extracting metabolites from the cell. Because of the relatively fast turnover of metabolites, the quenching and extraction efficiency become major hurdles in their measurements. In addition, metabolite measurements typically require liquid chromatography/mass spectrometry (LC/MS) or gas chromatography (GC)/MS, which may not be readily available in each laboratory, and the cost of instrumentation, is often prohibitive. This application proposes to solve these problems by developing a low cost, automated, chip-based sample preparation and HPLC system, which can be used in common laboratories next to the point-of-practice (either experimental or clinical). The HPLC output is fractionated by an on-chip fractionator, which is physically separated from, but ready to interface with, MS/MS in other laboratories. In this scheme, sample preparation is much more accurate and convenient because of automation; metabolite separation is more efficient because tandem HPLC can be build on chip; cost of measurement is significantly reduced because of the low cost of on-chip HPLC and outsourcing of the MS/MS operation to MS specialists.